Or LIBRA Jr.) to -1- I APOSTATIC SELECTION: THE RESPONSES OF WILD PASSERINES TO ARTIFICIAL POLYMORPHIC PREY by John Ashdown Allen Thesis submitted for the Degree of Doctor of Philosophy University of Edinburgh, December 1972 -2- ABSTRACT The iaintonanco of non-mimetic colour polymorphism has yet to be satisfactorily explained. Apotatic selection is one possible answer: sight-dependent predators may tend to form 'searching images' for common morphs and concentrate on those varieties in preference to rarer ones • (Chapter t). Some evidence for apostatic selection exists, but it is mostly indirect and incomplete'. (Chapter II). The hypothesis was tested by experiments with wild passerine birds in their normal surroundings. The prey wore lard-and-flour 'baits' which were usually green or brown and cylindrical (0.7cm long x 0.7cm diameter). Data were obtained by direct or indirect observation of 'populations' containing the colours in known proportions. Eaton baits were replaced frequently to ensure that the ratios wore kept constant. Five cots of experiments wore carried out. (chapter II!). In one set (Chapter IV) baits were at maximum density. Mon unequal numbers of greens and browns were presented there was a significant overall tendency for the ftrer forms to be removed. Individual blackbirds Tur'dua morula tended to concentrate on one colour alone within visits. Presentation of greens and browns in equal numbers gave no evidence that the colours differed in taste. Further experiments with small green and brown baits and small rod and yellow baits confirmed that rare forms are preferred at maximum density. -3- In the remaining sets of oxperimonte baits tiore at a density of tio per square metre. Populations of randomly distributed groans and browns were presented on grass or soil backgroncIe. A second sot of axperiments (Chapter V) revealed that birds can bocoize conditioned to searching for either green or broin baits. Mid birds were fed with one colour alone and toro then given populations with both colours in equal proportions. In all eight operiments the familiar colour was strongly preferred. In three cases the preferences woro reversed by a further training period on the'unfamiliar' colour. A third cot (Chapter VI) shoed that psscorinea on average tend to take more of the common variety than aEpected on the bcsi of constant selection • In each experiment populations tore presented either with greens nine times as comian as brome folloed by populations with brome nine times as common as greens or in the rerso order. The initial study revealed heterogeneity within the data and repeat experiments wora therofere carried Cut. The data frW the fourteen experiments wam consistent • In thirteen cases the common colour was preferred ovoll, despite different environmental conditions. A fourth sot investigated the behaviour of individual blackbirds and songtbr'uahos Turdua philociclos. One experiment (Chapter VII) showed that the three birds involved preferred brotmo • In two cases these preferences were very strong and the birds otartod to oat greens in quantity only tihen brww mere very rare • This behaviour woo preceded by an increase in the tendency to 'imprint-pew. The three birds appeared to differ markedly -'4 - in their rasponsas. Too further eporirents (Chapter VIII) with a majority of 'nc& birds confirmed the above findings and also provided evidence of searching images. At the and of the third experiment populations of green, browns and khstde were offered. The khakis were initially overlooked. In a fifth set (Chapter IX) greens, browns and various shades of khaki were used. Two series of experiments tooted the specificity of conditioning. Birds were familiarized with one colour and were than offered populations with the 'familiar' 11 bait and a visually similar colour in equal proportions. They were then trained on the 'unfamiliar' colour and given a choice as before. The results showed that birds can become conditioned to a particular colour; this conditioning can be very specific, but reversible • In two further experiments an attempt was made to see whether predators Can promote polyorphietm. Blackbirds and songthrusbes wore offered populations containing nine morphs ranging from green to brown and in proportions that fitted the normal distribution. These praportionQ were altered regularly to accord with the freuoncioa that z'oiaatnod after a fixed percentage of baits had been eaton • One eporiment showed that the browner baits were preferred. The second, experiment was preceded by a period of familiaritation with the cotxnzonoet khaki morph and showed an increase in the variance of the population. The distribution became bimodal • This disruptive effect was probably a result of the training, natural brown preferences, and the heterogeneity of behaviour of the birds.. Searching images can therefore be very specific and under certain conditions predators may be capable of causing polymorphism in prey species. -5-- The oxporixita1 dasign is conidosd to resemblo a natural situation. It is ocnàludod that the results support the hyothoio of apoetatic Ooli3ction • The findingo have Implications fbr the conditions under which spostatic selection is likely to act and point the way to futura rosGarch i It is ougostod that a no of different typea of opoetcxtie polymorphism odot. (Chapter X). -6 - PREFACE My aim in this work has been to test a hypothesis. It has been suggested that variability within a population of a palatable prey species can be maintained by 'epostatic selection' by predators, the meaning of which will become clear later. This posibility has been investigated by presenting populations of artificial prey to wild passer'ino birds in their normal surroundings. The exorcise is therefore oituatod in the area of zoology where animal behaviour and population genetics overlap. We are concerned with the behaviour of predators and more important, with its effect an variability in the prey population. The thesis is organised into ten Chziptorc. The first is introductory and provides the background to the problem. Existing evidence for apostatic selection is reviewed in the second Chapter. A description of the general experimental approach is given in the third and the experiments themselves are discussed in the following seven Chapters. Bulky data are tabulated in appendices at the ends of the relevant Chapters. The final Chapter focuses on the main conclusions and points out possible lines of future research. Acknowledgements First and foremost I thank my supervisor Professor Bryan Clarke for his constant encouragement, criticism and discussion throughout all stages of the work. I am also grateful to many other friends and colleagues who have discussed aspects of the work with me. In particular I must mention those members of the 'population genetics group' who were - 7 - at-the Department of Zoology, University of Edinburgh, between 1967 and 1971. I am indebted to Professor J.M. Mitchison and Professor P.M.B. Walker for the facilities that I have enjoyed in their Department. The research was supported by a Research Studentship from the Science Research Coulicil, to whom I extend my appreciation. Various other people gave help and I thank them all. They include those who allowed me to work on their land: Professor Bryan and Mrs. Ann Clarke; the late Mr. J. Ewing; my parents, Professor and Mrs. P. Allen; and the farmers of Midlothian who allowed ma access to their fields. I am especially grateful to Miss Nan Brownlie, Mr. Kim Howell, Mrs. A. Almeida and Mrs. A. Patel who typed the first draft, and in particular Mrs. D. M.Powell who typed the thesis in its present form. The final typing, assembling and binding was done in Reading while I was in Dar es Salaam and I am very much indebted to Mrs. Powell for handling the arrangements at that end. My wife, Eleanor, gave much appreciated assistance in checking the final typescript. Mr. Bob Brown photographed most of the Figures and Mr. David Briscoe photographed the Colour Plates. Finally, I thank a motley crowd of people without whose spiritual support this thesis would never have been finished. These include Eleanor, Boris and Chris, John Clarke, Bones, Cohn Craig, David and Jane, Captain Beefheart, and latterly Michael, John and Jan in Dar. All these people kept my sanity at times when prospects seemed bleak. -8- CONTENTS Page ABSTRACT 0 ,0 . 0 . • oo.o.... 000 000 2 PREFACE ..... 00006 ....................... 0 ..................6 Acknowledgements ........................... 6 TABLESTEXT -FIGURES, PLATES ............... , . ........... 13 TABLES. 0 0 0 0 00 0 0 ...... 00 ....................... 0 0 0 00 0 4 13 TEXT-FIGURES ........ 0 0 0 • , • • • . • • • ............... 15 3 • PLATES ........... • • • • • • 0 0 0 0 0 0 0 0 4 ..........16 CHAPTER I GENERAL INTRODUCTION .................. .......... 17 1. COLOUR POLYMORPHISM AND ITS MAINTENANCE ........ o.... 18 2. SELECTION BY PREDATORS , .......• 0 • • ................. 21 General 0 • • • • , • 0 .......• , • 0 0 0 0 0 0 0 0 0 0 0 0 • 0 0 00 0 • 0 0 21 Apostatic selection • • • • • • • •• • 00 0 0 0 0 24 3. APOSTATIC SELECTION AND SEARCHING IMAGES 26 Searching image 0 0 00 • • 0 • • • ...........0 • • • • ,0 26 Searching images and colour polymorphism 0 00 00 28 CHAPTER II THE EVIDENCE FOR APOSTATIC SELECTION ,.o....,, 29 1 . IN'rIDtJc'rIoN 0040000 • ..................., • 0 • • 30 2 . MIXED COLONIES OF CEPAEA ............................ 31 3. PREDATOR-PREY STUDIES ................. 0 ..........• • 32 (a)